The majority of the 1.5 million individuals who reside in our nation's nursing homes are elderly persons with dementias that place them at risk for poor health outcomes and diminished quality of life. Need driven behaviors, such as agitation and passivity, are exhibited by 90% of residents with dementia, and account for many poor health outcomes, including poor emotional functioning, decline in physical functioning, use of restraint, social isolation and increased risk of abuse. In the Need-driven Dementia compromised Behavior model (NDB) these behaviors are conceptualized as preserved remnants of earlier behavioral patterns that are triggered by unmet needs. Recreational activities tailored to the skill level and interests of residents hold promise as management strategies for agitation and passivity, but there has been no systematic examination of the treatment components (skill level and interest) that are hypothesized to result in therapeutic outcomes. Until recently, no theoretical framework that addressed nursing home residents' needs related to skill level and interests was available. The primary aim of this study is to test the efficacy of recreational activities derived from the NDB model in reducing agitation and passivity and improving engagement, affect and mood in nursing home residents with dementia. These recreational activities will be individually tailored to match the resident's skill level (cognitive ability and physical functioning) and interests (personality). The NDB-derived activities meet needs by addressing three important causes of need-driven behaviors: impaired cognitive ability, impaired physical functioning and premorbid personality. This study will be a randomized clinical trial with repeated measures, in which 140 residents will be randomly assigned to one of four groups for a 3-week intervention. Measures of agitation and passivity, engagement, affect and mood will be taken before and after interventions. It is hypothesized that residents who receive the NDB-derived intervention will exhibit significantly less agitation, less passivity, greater engagement, more positive affect and more positive mood than residents who receive the match to skill only, match to interest only, or control intervention. This study holds promise for improving quality of life for nursing home residents with dementia by using theoretically based activities to respond to their needs.